Tester
Lipp Laboratory, Department of Environmental Health Science, University of Georgia
2020-11-12
Load all required libraries.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 3.6.3## -- Attaching packages ---------------------------------------------------------------------------- tidyverse 1.3.0 --## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.3 v dplyr 1.0.0
## v tidyr 1.1.0 v stringr 1.4.0
## v readr 1.3.1 v forcats 0.5.0## Warning: package 'ggplot2' was built under R version 3.6.3## Warning: package 'tibble' was built under R version 3.6.3## Warning: package 'readr' was built under R version 3.6.3## Warning: package 'dplyr' was built under R version 3.6.3## Warning: package 'forcats' was built under R version 3.6.3## -- Conflicts ------------------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(plotly)## Warning: package 'plotly' was built under R version 3.6.3##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(broom)## Warning: package 'broom' was built under R version 3.6.3Read in raw data from RDS.
raw_data <- readRDS("./n1_n2_cleaned_cases.rds")Make a few small modifications to names and data for visualizations.
final_data <- raw_data %>% mutate(log_copy_per_L = log10(mean_copy_num_L)) %>%
rename(Facility = wrf) %>%
mutate(Facility = recode(Facility,
"NO" = "WRF A",
"MI" = "WRF B",
"CC" = "WRF C"))Seperate the data by gene target to ease layering in the final plot
#make three data layers
only_positives <<- subset(final_data, (!is.na(final_data$Facility)))
only_n1 <- subset(only_positives, target == "N1")
only_n2 <- subset(only_positives, target == "N2")
only_background <<-final_data %>%
select(c(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke)) %>%
group_by(date) %>% summarise_if(is.numeric, mean)
#specify fun colors
background_color <- "#7570B3"
seven_day_ave_color <- "#E6AB02"
marker_colors <- c("N1" = '#1B9E77',"N2" ='#D95F02')#remove facilty C for now
#only_n1 <- only_n1[!(only_n1$Facility == "WRF C"),]
#only_n2 <- only_n2[!(only_n2$Facility == "WRF C"),]
only_n1 <- only_n1[!(only_n1$Facility == "WRF A" & only_n1$date == "2020-11-02"), ]
only_n2 <- only_n2[!(only_n2$Facility == "WRF A" & only_n2$date == "2020-11-02"), ]Build the main plot
#first layer is the background epidemic curve
p1 <- only_background %>%
plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~new_cases_clarke,
type = "bar",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Daily Cases: ', new_cases_clarke),
alpha = 0.5,
name = "Daily Reported Cases",
color = background_color,
colors = background_color,
showlegend = FALSE) %>%
layout(yaxis = list(title = "Clarke County Daily Cases", showline=TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#renders the main plot layer two as seven day moving average
p1 <- p1 %>% plotly::add_trace(x = ~date, y = ~X7_day_ave_clarke,
type = "scatter",
mode = "lines",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Seven-Day Moving Average: ', X7_day_ave_clarke),
name = "Seven Day Moving Average Athens",
line = list(color = seven_day_ave_color),
showlegend = FALSE)
#renders the main plot layer three as positive target hits
p2 <- plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n1,
symbol = ~Facility,
marker = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n2,
symbol = ~Facility,
marker = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
layout(yaxis = list(title = "SARS CoV-2 Copies/L",
showline = TRUE,
type = "log",
dtick = 1,
automargin = TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#adds the limit of detection dashed line
p2 <- p2 %>% plotly::add_segments(x = as.Date("2020-03-14"),
xend = ~max(date + 10),
y = 3571.429, yend = 3571.429,
opacity = 0.35,
line = list(color = "black", dash = "dash")) %>%
layout(annotations = list(x = as.Date("2020-03-28"), y = 3.8, xref = "x", yref = "y",
text = "Limit of Detection", showarrow = FALSE))
p1## Warning: `arrange_()` is deprecated as of dplyr 0.7.0.
## Please use `arrange()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.## Warning: Ignoring 1 observations p2## Warning: `group_by_()` is deprecated as of dplyr 0.7.0.
## Please use `group_by()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.p_combined <-
plotly::subplot(p2,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationsp_combinedSave the plot to pull into the index
save(p_combined, file = "./plotly_fig.rda")Save an htmlwidget for website embedding
htmlwidgets::saveWidget(p_combined, "plotly_fig.html")Build loess smoothing figures figures
#create smoothing data frames
#n1
smooth_n1 <- only_n1 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N1")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#n2
smooth_n2 <- only_n2 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N2")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#add trendlines
#extract data from geom_smooth
#n1 extract
# *********************************span 0.6***********************************
#*****************Must always update the n = TOTAL NUMBER OF DAYS*************************
extract_n1 <- ggplot(smooth_n1, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n1<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 149)## Warning: Ignoring unknown aesthetics: outfit#n2 extract
extract_n2 <- ggplot(smooth_n2, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n2<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 149)## Warning: Ignoring unknown aesthetics: outfit#look at the fits to align dates and total observations
#n1
extract_n1## `geom_smooth()` using formula 'y ~ x'
fit_n1## [1] 11.43454 11.53574 11.63548 11.73299 11.82746 11.91814 12.00423 12.08495
## [9] 12.16106 12.23397 12.30382 12.37081 12.43509 12.49684 12.55624 12.61222
## [17] 12.66384 12.71144 12.75535 12.79592 12.83349 12.86839 12.90097 12.93156
## [25] 12.96050 12.98813 13.01480 13.04083 13.06657 13.08658 13.09718 13.10139
## [33] 13.10221 13.10265 13.10570 13.11438 13.11974 13.11223 13.09391 13.06687
## [41] 13.03318 12.99492 12.95417 12.91300 12.87350 12.83773 12.80778 12.78572
## [49] 12.77363 12.77359 12.78157 12.79195 12.80440 12.81861 12.83426 12.85103
## [57] 12.86861 12.89244 12.92705 12.97076 13.02187 13.07870 13.13955 13.20275
## [65] 13.26660 13.32941 13.38949 13.44517 13.49474 13.53651 13.56881 13.60999
## [73] 13.67331 13.74940 13.82893 13.90253 13.96086 13.99456 14.01074 14.02289
## [81] 14.03131 14.03626 14.03801 14.03684 14.03302 14.02683 14.01853 14.00841
## [89] 13.99673 13.98376 13.96979 13.95509 13.92957 13.88699 13.83333 13.77460
## [97] 13.71679 13.66588 13.62789 13.59401 13.55242 13.50452 13.45175 13.39551
## [105] 13.33723 13.27833 13.22022 13.16433 13.11207 13.06487 13.02414 12.99130
## [113] 12.96777 12.95537 12.95279 12.95675 12.96398 12.97120 12.97515 12.97256
## [121] 12.96696 12.96396 12.96334 12.96488 12.96839 12.97364 12.98043 12.98853
## [129] 12.99775 13.00787 13.01868 13.02996 13.04151 13.05311 13.06572 13.08019
## [137] 13.09616 13.11323 13.13105 13.14922 13.16792 13.18754 13.20810 13.22958
## [145] 13.25201 13.27537 13.29967 13.32337 13.34681#n2
extract_n2## `geom_smooth()` using formula 'y ~ x'
fit_n2## [1] 11.19880 11.34172 11.48206 11.61907 11.75199 11.88006 12.00252 12.11862
## [9] 12.22909 12.33528 12.43736 12.53547 12.62978 12.72045 12.80762 12.89025
## [17] 12.96741 13.03945 13.10669 13.16949 13.22818 13.28310 13.33460 13.38302
## [25] 13.42870 13.47198 13.51320 13.55270 13.59083 13.62262 13.64470 13.65975
## [33] 13.67047 13.67954 13.68967 13.70354 13.71157 13.70393 13.68301 13.65118
## [41] 13.61079 13.56423 13.51385 13.46203 13.41114 13.36355 13.32162 13.28772
## [49] 13.26422 13.25350 13.24648 13.23422 13.21924 13.20407 13.19125 13.18330
## [57] 13.18276 13.19156 13.20908 13.23397 13.26492 13.30060 13.33967 13.38082
## [65] 13.42271 13.46402 13.50341 13.53956 13.57115 13.59683 13.61530 13.63912
## [73] 13.67772 13.72505 13.77504 13.82163 13.85877 13.88040 13.89172 13.90194
## [81] 13.91103 13.91893 13.92560 13.93099 13.93506 13.93775 13.93903 13.93884
## [89] 13.93713 13.93387 13.92901 13.92249 13.90749 13.88010 13.84465 13.80545
## [97] 13.76683 13.73312 13.70865 13.68943 13.66873 13.64671 13.62353 13.59935
## [105] 13.57433 13.54864 13.52244 13.49588 13.46914 13.44237 13.41573 13.38939
## [113] 13.36351 13.33437 13.29970 13.26184 13.22311 13.18585 13.15238 13.12504
## [121] 13.10130 13.07722 13.05296 13.02872 13.00467 12.98100 12.95789 12.93551
## [129] 12.91406 12.89371 12.87464 12.85704 12.84109 12.82697 12.81424 12.80235
## [137] 12.79124 12.78091 12.77131 12.76241 12.75420 12.74671 12.73997 12.73401
## [145] 12.72887 12.72457 12.72115 12.71899 12.71799#assign fits to a vector
n1_trend <- fit_n1
n2_trend <- fit_n2
#extract y min and max for each
limits_n1 <- ggplot_build(extract_n1)$data## `geom_smooth()` using formula 'y ~ x'limits_n1 <- as.data.frame(limits_n1)
n1_ymin <- limits_n1$ymin
n1_ymax <- limits_n1$ymax
limits_n2 <- ggplot_build(extract_n2)$data## `geom_smooth()` using formula 'y ~ x'limits_n2 <- as.data.frame(limits_n2)
n2_ymin <- limits_n2$ymin
n2_ymax <- limits_n2$ymax
#reassign dataframes (just to be safe)
work_n1 <- smooth_n1
work_n2 <- smooth_n2
#fill in missing dates to smooth fits
work_n1 <- work_n1 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n1 <- work_n1$date
work_n2 <- work_n2 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n2 <- work_n2$date
#create a new smooth dataframe to layer
smooth_frame_n1 <- data.frame(date_vec_n1, n1_trend, n1_ymin, n1_ymax)
smooth_frame_n2 <- data.frame(date_vec_n2, n2_trend, n2_ymin, n2_ymax)#make plotlys
#plot smooth frames
p3 <- plotly::plot_ly() %>%
plotly::add_lines(x = ~date_vec_n1, y = ~n1_trend,
data = smooth_frame_n1,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1,
'</br> Median Log Copies: ', round(n1_trend, digits = 2),
'</br> Target: N1'),
line = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_lines(x = ~date_vec_n2, y = ~n2_trend,
data = smooth_frame_n2,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2,
'</br> Median Log Copies: ', round(n2_trend, digits = 2),
'</br> Target: N2'),
line = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_ribbons(x ~date_vec_n1, ymin = ~n1_ymin, ymax = ~n1_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1, #leaving in case we want to change
'</br> Max Log Copies: ', round(n1_ymax, digits = 2),
'</br> Min Log Copies: ', round(n1_ymin, digits = 2),
'</br> Target: N1'),
name = "",
line = list(color = '#1B9E77')) %>%
plotly::add_ribbons(x ~date_vec_n2, ymin = ~n2_ymin, ymax = ~n2_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2, #leaving in case we want to change
'</br> Max Log Copies: ', round(n2_ymax, digits = 2),
'</br> Min Log Copies: ', round(n2_ymin, digits = 2),
'</br> Target: N2'),
name = "",
line = list(color = '#D95F02')) %>%
layout(yaxis = list(title = "Total Log SARS CoV-2 Copies",
showline = TRUE,
automargin = TRUE)) %>%
layout(xaxis = list(title = "Date")) %>%
plotly::add_segments(x = as.Date("2020-06-24"),
xend = as.Date("2020-06-24"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Bars Repoen",
hoverinfo = "text",
text = "</br> Bars Reopen",
"</br> 2020-06-24",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-07-09"),
xend = as.Date("2020-07-09"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Mask Mandate",
hoverinfo = "text",
text = "</br> Mask Mandate",
"</br> 2020-07-09",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-08-20"),
xend = as.Date("2020-08-20"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> Classes Begin",
"</br> 2020-08-20",
hoverinfo = "text",
text = "Classes Begin",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-10-03"),
xend = as.Date("2020-10-03"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> First Home Football Game",
"</br> 2020-10-03",
hoverinfo = "text",
text = "First Home Football Game",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n1,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#1B9E77', size = 6, opacity = 0.65)) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n2,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#D95F02', size = 6, opacity = 0.65))
p3Create final trend plot by stacking with epidemic curve
smooth_extract <-
plotly::subplot(p3,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationssmooth_extractsave(smooth_extract, file = "./smooth_extract.rda")